General: Cold bridge-free thresholds for heavy sliding/bifold doors

1 to 30 of 42

- CommentAuthorlineweight
- CommentTimeMar 4th 2019
Can anyone point me to some good details showing a good way to deal with the threshold of a heavy sliding door, without creating a cold bridge?

Generally they want a fully supported solid base at least the depth of the window frame itself, which seems to mean concrete/masonry spanning from inside to out. I previously tried to deal with this in a primarily timber framed design using a timber joist/ply box type arrangement but in the end the installers didn't like it, and demanded tolerances for movement/deflection which the engineer said he couldn't achieve in timber. We ended up using aerated blocks, with perimeter insulation to the floor screed turned up behind it.

Now I'm looking at one where we have a conventional cavity wall, and suspended floor. The base of the sliders will need to bridge the cavity so whatever supports it will also bridge the insulation layer. Is there a best way of doing this?
- CommentAuthorbhommels
- CommentTimeMar 4th 2019 edited
Not sure how much help it is to you, but just to describe my case where a triple glazed french door set had to sit sort of halfway a 200mm cavity:
Along the width of the door, the inner leaf of the cavity wall was lowered and moved into the cavity by 120ish mm. This lowered wall sits on the concrete cavity infill, quite a long way down in the cavity, which is filled with XPS otherwise.
Compacfoam sits on top of this lowered wall, with XPS filling the gap between the compacfoam and outer leaf. Celotex was used to fill the gap on the inside and provide perimeter insulation. Studs & bolts through the compacfoam, XPS and outer leaf hold the lot in place. The doors sit on the compacfoam, I have not had any problems with movement so far, and the thermal bridging is not too bad.
- CommentAuthorgyrogear
- CommentTimeMar 4th 2019
When I asked my installers about this, they said that the (UPVC) frame is in and of itself anti-cold-bridging due to the construction...

So it sits directly half-on and half-off of the concrete slab (I wanted it to sit on 20mm XPS). They said they could not do this, as it had been sized and built for the net dimensions of the aperture.

Therefore (in my view) the lesson is that the pre-job measuring-up was wrong.

This was not the only dropped blk on my install... (I found out afterwards that the sales engineer later got fired).

gg
- CommentAuthorlineweight
- CommentTimeMar 4th 2019
Posted By: bhommelsNot sure how much help it is to you, but just to describe my case where a triple glazed french door set had to sit sort of halfway a 200mm cavity:
Along the width of the door, the inner leaf of the cavity wall was lowered and moved into the cavity by 120ish mm. This lowered wall sits on the concrete cavity infill, quite a long way down in the cavity, which is filled with XPS otherwise.
Compacfoam sits on top of this lowered wall, with XPS filling the gap between the compacfoam and outer leaf. Celotex was used to fill the gap on the inside and provide perimeter insulation. Studs & bolts through the compacfoam, XPS and outer leaf hold the lot in place. The doors sit on the compacfoam, I have not had any problems with movement so far, and the thermal bridging is not too bad.

Thanks.

I should have a look at Compacfoam.

I guess the depth of your door frame was not greater than the width of the inner leaf, so it could sit entirely on that?

The issue with bifolds - and sliders particularly - is that the frame can be very deep, and wider than a single blockwork leaf. In the instance I'm looking at just now, for example, it's about 160mm deep, hence the need to span the insulation layer. That's for a 2-panel slider but it is even greater for a 3 or more panel slider of course.

A supplier I just spoke to now said their preferred approach is simply to fill the cavity with concrete, over the portion of the wall where the opening is. Then deal with insulation by perimeter insulation around the floor screed (if there is one). If it's a suspended timber floor, then I guess you just have to rely on the between-joist insulation where it meets the wall and the back of the glazing frame (which is set just below finished floor level).
- CommentAuthorlineweight
- CommentTimeMar 4th 2019
Posted By: gyrogearWhen I asked my installers about this, they said that the (UPVC) frame is in and of itself anti-cold-bridging due to the construction...

So it sits directly half-on and half-off of the concrete slab (I wanted it to sit on 20mm XPS). They said they could not do this, as it had been sized and built for the net dimensions of the aperture.

Therefore (in my view) the lesson is that the pre-job measuring-up was wrong.

This was not the only dropped blk on my install... (I found out afterwards that the sales engineer later got fired).

gg

Is the implication that it could have sat on XPS if the sizing had been right? What sort of weight was it?
- CommentAuthorowlman
- CommentTimeMar 4th 2019
The weight issue could depend on whether the Bi-folds are top hung or bottom rolling; they can be either or both.
- CommentAuthorlineweight
- CommentTimeMar 4th 2019 edited
Posted By: owlmanThe weight issue could depend on whether the Bi-folds are top hung or bottom rolling; they can be either or both.

Bottom rolling in my case.
- CommentAuthorbhommels
- CommentTimeMar 4th 2019 edited
Posted By: lineweight
I should have a look at Compacfoam.

I guess the depth of your door frame was not greater than the width of the inner leaf, so it could sit entirely on that?

The issue with bifolds - and sliders particularly - is that the frame can be very deep, and wider than a single blockwork leaf. In the instance I'm looking at just now, for example, it's about 160mm deep, hence the need to span the insulation layer. That's for a 2-panel slider but it is even greater for a 3 or more panel slider of course.

A supplier I just spoke to now said their preferred approach is simply to fill the cavity with concrete, over the portion of the wall where the opening is. Then deal with insulation by perimeter insulation around the floor screed (if there is one). If it's a suspended timber floor, then I guess you just have to rely on the between-joist insulation where it meets the wall and the back of the glazing frame (which is set just below finished floor level).

The frame depth for our door set is 70-80mm. The reason for siting it halfway the cavity insulation is to get the best thermal performance.
Filling the cavity with concrete to just under DPC is common for some builders. It turns your walls into a massive heat sink unfortunately: better to have an infill at very low level only (if at all) and fill the rest with snug fitting XPS.

I would not rely on XPS300 to suspend a bifold door set. Compacfoam is thermally 3x worse than XPS, but has 14x the compression strength.
(lambda of 0.09 W/m2K, 1900kPa (long-term) for compacfoam against 0.033 and 130 for XPS300).
- CommentAuthorgyrogear
- CommentTimeMar 4th 2019 edited
Posted By: lineweightIs the implication that it could have sat on XPS if the sizing had been right? What sort of weight was it?

Yes. I don't know the weight: they are not actually doors but large-size DG dual pull-and-slide windows, 1400 square - they form the south-facing element of a corner glazing system. The overhanging half of the frame sits over the EWI. I instinctively did not want the uPVC touching the slab - after all the side rails do not touch the walls !

If guessing, I'd say each weighs around 120 kgs, therefore 240 kgs total, with the weight distributed over a rectangle of 280 x 5 cm, so I guess that makes a load of 0.0.17 kg/cm2 or 2.35 lbs/sq.inch if my sums are right...

PSX seems to take around 40 lbs/sq.in.

gg
- CommentAuthorlineweight
- CommentTimeMar 4th 2019
Going with the supplier-suggested default of filled cavity, and a suspended timber floor with insulation between the joists, looks something like this drawn to scale (attached).

We are dealing with aluminium frames here (yes I know not great, but that is what has been chosen).

The cold bridge through the wall is indicated by the large arrow. I can see that a layer of something like compacfoam would improve this by moving the right-hand end of the arrow downwards, further into the depth of the between-joist insulation.

But it doesn't do anything about the heat path through the aluminium frame. The most that can be done about that, I think, is a small strip of perimeter insulation between it and the edge of the floor finishes but it's so small it seems it would have no major effect.

It makes me think, if we are accepting aluminium frames then is worrying about cold bridge through whatever's underneath them a bit pointless?

This is where I wish I had and could use Therm-type software to quantify what difference it would make.
- CommentAuthorgyrogear
- CommentTimeMar 4th 2019 edited
Posted By: lineweightI can see that a layer of something like compacfoam would improve this

yes - or even 20mm XPS or less... or a strip of neoprene.

The aluminium downstands would probably CUT into the XPS, but at least latter would then fill that little void (which to me looks like another thermal bridge over and above your red arrow).

OR drill a hole in the bottom of the sill and inject with expanding foam ?
gg
- CommentAuthorbhommels
- CommentTimeMar 4th 2019 edited
The aluminium is not great, on the other hand the cross section is only small. I would worry about condensation on the inner leaf with all that timber nearby, which a block of compacfoam would also nicely do away with.
The attached picture is what I ended up with, save for the cill/outer leaf detail, which has been improved.
- CommentAuthorlineweight
- CommentTimeMar 4th 2019
Posted By: gyrogear
Posted By: lineweightI can see that a layer of something like compacfoam would improve this

yes - or even 20mm XPS or less... or a strip of neoprene.

The aluminium downstands would probably CUT into the XPS, but at least latter would then fill that little void (which to me looks like another thermal bridge over and above your red arrow).

OR drill a hole in the bottom of the sill and inject with expanding foam ?
gg

That void actually has a kind of insulated block set into it that also incorporates the drainage routes from the slider tracks. So I imagine is designed with the intention that there is no weight going through it, which would be one reason not to try and sit it on anything that was going to compress. This I imagine is partly why they say they need a solid base. So I don't think XPS would work.
- CommentAuthorlineweight
- CommentTimeMar 4th 2019
Posted By: bhommelsThe aluminium is not great, on the other hand the cross section is only small. I would worry about condensation on the inner leaf with all that timber nearby, which a block of compacfoam would also nicely do away with.
The attached picture is what I ended up with, save for the cill/outer leaf detail, which has been improved.
Yes, that is how they sell this type of alu frame, by saying the cross section is small and in some cases encouraging insulating right up to the frame.

I may see how they react to the suggestion of installing onto something like compacfoam. By the way a lambda of 0.09 doesn't seem hugely better than aerated block that I think can be around 0.11; does it have any other advantage?
- CommentAuthorbhommels
- CommentTimeMar 4th 2019 edited
It is homogenous and easy to cut to size, and comes in 1 m lengths of 150x100mm cross section max. The lambda is actually listed as 0.087, so still 25% better than the best aerated block. Presumably much more expensive though.
- CommentAuthorMikC
- CommentTimeMar 4th 2019
Alternatives to compact foam might be foamglas perinsul

https://uk.foamglas.com/en-gb/products/product-overview/foamglas-perinsul/foamglas-perinsul-hl-high-load

â‰¤ 0,058 W/(m.K)

Just a thought..
- CommentAuthorbhommels
- CommentTimeMar 5th 2019
Posted By: MikCAlternatives to compact foam might be foamglas perinsul

https://uk.foamglas.com/en-gb/products/product-overview/foamglas-perinsul/foamglas-perinsul-hl-high-load" rel="nofollow" >https://uk.foamglas.com/en-gb/products/product-overview/foamglas-perinsul/foamglas-perinsul-hl-high-load

â‰¤ 0,058 W/(m.K)

Just a thought..

Good find! Looks perfectly suitable for this kind of application.
- CommentAuthordjh
- CommentTimeMar 5th 2019 edited
Posted By: gyrogearyes - or even 20mm XPS or less... or a strip of neoprene.

I don't think either of those would be man enough for the job.

Compacfoam or Marmox Thermoblocks or the foamglas product or similar would be better suited I think.

Colin, doesn't that cross-section you posted show that there are thermal breaks in the aluminium? It's not a solid piece of aluminium I don't think. What is the vertically dashed material underneath it?
- CommentAuthorlineweight
- CommentTimeMar 6th 2019
Posted By: djh
Posted By: gyrogearyes - or even 20mm XPS or less... or a strip of neoprene.

I don't think either of those would be man enough for the job.

Compacfoam or Marmox Thermoblocks or the foamglas product or similar would be better suited I think.

Colin, doesn't that cross-section you posted show that there are thermal breaks in the aluminium? It's not a solid piece of aluminium I don't think. What is the vertically dashed material underneath it?

It is thermally broken yes. As I understand it though, even thermally broken aluminium is never all that great.

The material underneath it, not sure exactly what it is but on the drawings I have it's labelled as 'insulation panel for drainage'.
- CommentAuthorlineweight
- CommentTimeMar 6th 2019
By the way here's a detail I've been given for another system we're looking at, which claims a better Uw value.

NB what is described as 'Keller Base Stone'. I've not got as far as finding out what that is yet; not sure if it's an insulated block of some kind.
- CommentAuthorgoodevans
- CommentTimeMar 6th 2019
Door threshold details is where I had to compromise - I don't think you can drill into the marmox or foamglass products they are not homogeneous. I think compacfoam is and can be cut/drilled as if it was wood.
- CommentAuthorlineweight
- CommentTimeMar 26th 2019
Coming back to this... further investigations brought me to "structural cavity closers" and products from Cavalok who seem to be part of Eurocell.

This has been mentioned in a previous thread, I see:

http://www.greenbuildingforum.co.uk/newforum/comments.php?DiscussionID=9392

It's hard to find any very useful information on the Eurocell site. I had a bit of a confusing conversation with one of their technical people, where I tried to explain that I wanted to understand whether they can offer something that will span a cavity under the sill and take the weight of a door/window installed above it. His answer seemed to be that this would never be an issue because the frame itself can simply span the cavity, but that's not what the window installers I've talked to have said. They always want a continuous solid base across the depth of the frame.

Anyway - they pointed me to here:

https://www.subframes.co.uk/cavity-closers/structural-door-packs/

And I'm making some enquiries with them. But I think that this system is not actually designed to be load bearing across the bottom horizontal member - it's more that you can fix directly into the jamb closers and they will transfer the load into the outer and inner leaves.
- CommentAuthorTimSmall
- CommentTimeMar 27th 2019
Compacfoam comes in a variety of compressive strengths. The higher strength versions have lower insulation value. I used Compacfoam 150 under my sliding doors (around 200 kg for the sliding leaf!).

The one that's most widely available in the UK is "Compacfoam 200", which is 0.046 W/m.K.

In my experience XPS could be used under most sliding doors. You just have to make a reasonable estimate of the area under each roller which is bearing the weight, and how much weight each is taking, then compare that with the SWL for each material.

We used a power plane to level the Compacfoam precisely so that the sill sat straight on it, and had continuous contact. The CF was bedded on mortar beforehand. You could probably do the same with higher density XPS (or EPS) sheets.

Looking at that sill profile, you could just bed it on mortar, but have some inserts of celotex to sit under the thermal breaks in the sill (the blue bits in the diagram above). You could glue down the celotex and use it as a "former" to place mortar between (leveling the celotex with a rasp).
- CommentAuthorlineweight
- CommentTimeMar 27th 2019 edited
Posted By: TimSmall
Looking at that sill profile, you could just bed it on mortar, but have some inserts of celotex to sit under the thermal breaks in the sill (the blue bits in the diagram above). You could glue down the celotex and use it as a "former" to place mortar between (leveling the celotex with a rasp).

Makes sense - I can see a regular general builder raising his eyebrows at something like this though!
- CommentAuthorDoubting_Thomas
- CommentTimeMar 27th 2019
+1 for Compacfoam.

I used this in my house for some Internorm sliding doors and it worked fine. The most difficult aspect was finding stainless steel screw ties to cast into the slab so that it doesn't move. Plus dressing the DPM around the compacfoam block can be fiddly.

The image below is from the Green Building Store who are normally quite good at sharing knowledge on this kind of thing if you want more details.
https://www.greenbuildingstore.co.uk/products/compacfoam-200/
- CommentAuthorlineweight
- CommentTimeMar 27th 2019
^ Thanks for that.

Looking at the drawing I didn't really understand why the compactfoam is sitting onto the EPS. If the EPS can take the weight of door plus compacfoam then why not sit the door straight onto the EPS (it doesn't look like the compacfoam block spreads the weight out onto a larger area).

Then I looked at the other on GBS website (below)

Which clarifies that it's bolted into the side of the conc slab on the threaded rods. So does that mean that it's effectively a kind of cantilever and the weight is transferred back to the slab? Intuitively it seems improbable that that would work.
- CommentAuthorDoubting_Thomas
- CommentTimeMar 27th 2019
Yes, if feels odd but the way someone explained it to me: the rigid insulation (at least in my case) is taking the whole weight of the slab elsewhere so it should be capable of the load.

What you don't want is flex in the door frame. The threaded bars give you rigidity by tying back to the slab. It's effectively the best of both worlds in that you get the stability of a concrete slab, without the cold bridging issues of sitting something directly onto the slab.
- CommentAuthorlineweight
- CommentTimeMar 27th 2019
Posted By: Doubting_ThomasYes, if feels odd but the way someone explained it to me: the rigid insulation (at least in my case) is taking the whole weight of the slab elsewhere so it should be capable of the load.

What you don't want is flex in the door frame. The threaded bars give you rigidity by tying back to the slab. It's effectively the best of both worlds in that you get the stability of a concrete slab, without the cold bridging issues of sitting something directly onto the slab.

Is it offered as a 'system' by GBS? In other words if you give them the dimensions, weight of your door frame etc can they confirm whether this detail will work?

(Doesn't actually help me with the detail I'm currently looking at because we have a suspended timber floor, no concrete slab. Useful to remember for the future though.)
- CommentAuthordjh
- CommentTimeMar 27th 2019
Posted By: lineweightDoesn't actually help me with the detail I'm currently looking at because we have a suspended timber floor, no concrete slab. Useful to remember for the future though.

Well, you could fasten the compacfoam to whatever forms the edge of your timber floor structure. I presume there's some kind of perimeter beam. Might need to check the engineering - double up the beam or something in that area.
- CommentAuthorDoubting_Thomas
- CommentTimeMar 28th 2019 edited
Posted By: lineweightIs it offered as a 'system' by GBS? In other words if you give them the dimensions, weight of your door frame etc can they confirm whether this detail will work?

I don't think so, I'm afraid.

From memory I ended up doing some rough calculations that others on here helped out with. See second half of this thread: http://www.greenbuildingforum.co.uk/newforum/comments.php?DiscussionID=11503&page=1

The safety factor was big enough that I felt comfortable using it. Plus I also checked it with the supplier of the Isoquick and the Sliding Doors and it's a fairly standard approach.

1 to 30 of 42



	Site Map \| Home \| View Cart \| Pressroom \| Business \| Links	Contact Us
Logout
© Green Building Press

Green Building Forum

Categories

Feeds

General: Cold bridge-free thresholds for heavy sliding/bifold doors